Within the field of telecommunication systems, technologies for creating so called ad-hoc networks have been developed in order to for example simplify network deployment and provide a more dynamic network environment. An ad-hoc network comprises a set of network nodes, such as cellular phones. A network node may join or leave the ad-hoc network for various reasons, such as mobility, coverage conditions and more. In order for the network node to be able to join the ad-hoc network, a joining network node, i.e. a network node that wishes to join the ad-hoc network, is required to be able to discover the ad-hoc network. This means that the ad-hoc network comprises the network nodes that have already joined the ad-hoc network. Moreover, the joining node is required to be able to be discovered by the nodes in the ad-hoc network. This is generally referred to as discovery, and in particular to Device-to-Device (D2D) discovery.
Hence, in ad-hoc networking, D2D discovery refers to a procedure that allows devices in the proximity of each other to detect the presence of one another. Sometimes, D2D discovery may be referred to as neighbor or peer discovery. D2D discovery in ad hoc networks involves an engineering tradeoff between energy efficiency, discovery range, the number of discovered devices and the discovery time. Typically, ad hoc technologies such as Bluetooth deal with this problem by carefully designing measurement and beacon signaling procedures and employing state transitions between energy conserving and active states, e.g. between a beacon detecting state and a beacon transmitting state.
D2D discovery is a well-known and widely used component of many existing wireless technologies, including ad hoc and cellular networks. Apart from Bluetooth, several variants of the IEEE 802.11 standards suite, such as WiFi Direct include technical solutions for device discovery. A key technique used by these standards is to use specially designed so called beacon signals that devices can broadcast and capture, so that nearby devices are able to detect the proximity as well as some characteristics of such beacon broadcasting devices.
Beacon signaling based neighbor device discovery requires that a broadcasting device and a receiving device meet in the time, frequency and code domains. Furthermore, in order for discovery to work, the beacon receiving device is required to be able to decode the information encoded in the beacon signal. In other words, the beacon signal must reach a certain Signal-to-Noise-and-Interference Ratio (SINR) threshold at the receiving device in order for the beacon signal to be detected at the receiving device.
Although D2D discovery for ad hoc networks, using technologies such as Bluetooth, WiFi Direct, has been known for some time, only few techniques have been proposed and built for devices operating in cellular spectrum and using cellular technologies with or without the need for a base station. Hence, in the context of D2D discovery techniques for devices operating in cellular spectrum and using cellular technologies a few shortcomings have been identified.
A first known technique uses random selection of Peer Discovery Resources (PDR), which is a set of resources selected from within the cellular spectrum. The PDRs are dedicated for beacon signaling. With this technique, each beacon broadcasting device randomly selects one PDR, or as many as required for the beacon signal to be broadcast, out of dedicated PDRs within the cellular time-frequency resources. This technique may lead to situations in which multiple devices in the vicinity of each other use colliding PDRs. Such collisions of beacon signals make beacon signals undetectable or not decodable by receiving devices. Thus, D2D discovery by means of the beacon signals becomes difficult or even impossible.
A second known technique uses so called greedy, or opportunistic, selection of PDRs. This means that a device selects a PDR that is not used. When the number of nodes in the ad-hoc network exceeds the number of available discovery resources, all PDRs are being used at a given point in time. In this case, when a new node joins the ad-hoc network, or a device population, the new node selects one PDR for which a distance to the closest peer, or neighbor device, currently using that resource is maximized. This may be expressed by:
      C    ⁡          (      i      )        =      arg    ⁢                  ⁢                  max        k            ⁢              (                                            min                              j                ∈                                  N                  k                                                      ⁢                          (                                                                                    X                    i                                    -                                      X                    j                                                                              )                                ,                    where C(i) denotes the selected resource for node-i, and ∥Xi−Xj∥ denotes the Euclidean distance between node-i and node-j. Nk is the set of the nodes that are assigned to the resource k when node i joined the network. Selecting the resource according to the above formula corresponds to so called greedily picking a resource, sometimes referred to as channel, that produces the best Euclidean separation between the node selecting the peer discovery resource and other peers using the same peer discovery resource. A first noteworthy characteristic of this distance based peer discovery resource selection algorithm is that it requires the capability of performing measurements per peer discovery resource and per neighbor node. A second noteworthy characteristic of the distance based selection objective above is that it involves minimizing the distance to a particular peer. In certain scenarios, alternative techniques of separating PDRs selected by different nodes may be required.